The proposed research calls for three interviews at yearly intervals with a sample of 1,200 people 65 years and older residing in three Maryland counties. Its general aim is to account for the association between people's socioeconomic statuses (SES) and their health and well-being. In addition to age, these statuses include economic level, education, occupational prestige, marital status, gender, and race and ethnicity. The study is guided by the perspectives of the stress process conceptual framework. The specific aims of the inquiry essentially involve empirical tests of several interrelated assumptions underlying this conceptualization. Thus, it is assumed that SES potentially influences the nature and intensity of stressors to which people are exposed. Certain types of stressors, such as loss events, can become particularly prevalent in late life. Moreover, it is thought that exposure to one or more serious stressors may generate exposure to additional stressors, a phenomenon referred to as stress proliferation. Exposure to both initial and proliferated stressors, in turn, can lead to a variety of physical and mental health outcomes; in this way, stressful life conditions stand as one of the mechanisms linking SES and health. However, exposure to stressors does not always have a deleterious effect on health. It is proposed that even under difficult life conditions certain kinds of social and personal resources, such as social support and the sense of mastery, can protect individuals from these effects. It is further proposed that the particular health outcomes that are manifested under conditions of stress will vary with the SES characteristics of people.